EpiMCI: Predicting Multi-Way Chromatin Interactions from Epigenomic Signals

SIMPLE SUMMARY: Dissecting the relationship between epigenome signals and three-dimensional multi-way chromatin interactions remains a challenging problem. The emergence of high-throughput Pore-C technology offers promising hope for tackling this issue. In this study, we proposed the EpiMCI, a framework based on a hypergraph neural network, aiming to reconstruct multi-way chromatin interactions from epigenomic signals. The model obtained AUCs of 0.981 and 0.984 using the GM12878 and K562 datasets, outperforming the existing methods. The EpiMCI can be used to denoise multi-way contact sequencing data and improve data quality. The embeddings obtained from the EpiMCI reflect the exact genome structure, confirming the rationality of the EpiMCI from a biological perspective. Thus, the EpiMCI is a promising framework for reconstructing multi-way chromatin interactions from epigenomic signals and can be applied to studies related to multi-way chromatin interactions reconstruction. ABSTRACT: The recently emerging high-throughput Pore-C (HiPore-C) can identify whole-genome high-order chromatin multi-way interactions with an ultra-high output, contributing to deciphering three-dimensional (3D) genome organization. However, it also brings new challenges to relevant data analysis. To alleviate this problem, we proposed the EpiMCI, a model for multi-way chromatin interaction prediction based on a hypergraph neural network with epigenomic signals as the input. The EpiMCI integrated separate hyperedge representations with coupling hyperedge information and obtained AUCs of 0.981 and 0.984 in the GM12878 and K562 datasets, respectively, which outperformed the current available method. Moreover, the EpiMCI can be applied to denoise the HiPore-C data and improve the data quality efficiently. Furthermore, the vertex embeddings extracted from the EpiMCI reflected the global chromatin architecture accurately. The principal component analysis suggested that it was well aligned with the activities of genomic regions at the chromatin compartment level. Taken together, the EpiMCI can accurately predict multi-way chromatin interactions and can be applied to studies relying on chromatin architecture.